software(web) craftsman • oss • onchain • occasional author

westside africa
🚨✨ We’ve just lunched our Public Beta! Over the past few months, I and @OfuzorEmeke have been building OBVERSE in stealth An onchain + social invoicing for modern payments, create payment links, get paid onchain, and make transactions social. ˗ˏˋ obverse.cc ˎˊ-
🔥We made this video to show > not just tell > what drives us at OBVERSE. 🎥 Video dropped → cinematic explanation of OBVERSE in full effect. chat-based stablecoin invoicing magic. No friction. Pure flow. THIS IS US! obverse.cc
3
15
52
3,709
Vinyl Davyl 🪽 retweeted
checked them out. their own “context handoff” is model-to-model inside Pi (anthropic -> openAI in the same chat). not like claude -> cursor found a could but don’t really like them
Replying to @Vinylchi
Check out @pidotdev they have an extension that exports sessions between devices and even among users
2
4
63
how are people handling context transfer between AI agents today? JSON? mcp? state stores? shared memory? or something else. haven’t still found something that suits my taste
4
2
4
69
Vinyl Davyl 🪽 retweeted
how are people handling context transfer between AI agents today? JSON? mcp? state stores? shared memory? or something else. haven’t still found something that suits my taste
4
2
4
69
Vinyl Davyl 🪽 retweeted
bangerlistic. i came for this lectures at your place but you gave me a PS5 pad ehn? that thing no pure :( start crash course, i would be the first to register
Ok so many people resonated with this.
1
1
3
230
Vinyl Davyl 🪽 retweeted
If you’ve ever fought with JSON Schema, UML, or scattered data models across teams! 🧱 Checkout “Concerto” lets you write one clear model, get validated TypeScript, Java, Go, GraphQL & more. Powering smart legal contracts at scale 🔗 concerto.accordproject.org
2
4
117
love the “capability gaslighting” chat in this talk. frontier models would convince dawgs they’re an expert at some but fail the same task the next day.
What can us, engineers else learn from designers and design engineers? Turns out there’s lots, based on talking with one of the best design engineers in the industry, @Mappletons. Timestamps: 00:00 Intro 03:24 From anthropology to tech 10:18 What does a designer do? 18:23 How Maggie works 24:55 The case for planning with physical tools 31:53 Why Maggie is learning woodworking 33:13 Design engineers and engineering constraints 38:49 How Maggie uses Figma 40:30 Design at GitHub Next 45:12 How has AI changed design 50:37 When models design and why humans are still needed 53:30 UX and UI 58:29 Capability gaslighting 1:00:33 One Developer, Two Dozen Agents, Zero Alignment 1:07:21 Craft and AI tells 1:14:17 Visual gardens, home-cooked software, and barefoot developers 1:21:02 Advice for engineers and lessons from anthropology 1:25:34 Book recommendation Brought to you by: • @turbopuffer – a vector and full-text search engine built on object storage. It’s fast, cheap, and extremely scalable. turbopuffer.com/pragmatic • @AntithesisHQ – verify your system’s correctness without human review or traditional integration tests – and avoid bugs or outages. antithesis.com/pragmatic • @EntireHQ – every agent prompt, tool call, stored in your repo, and mirrored. Set up with one click: entire.io/pragmatic Hope you enjoy this unusually visual episode - thanks again, Maggie!
1
2
4
175
gtp-class models seems to lead race here
22
Vinyl Davyl 🪽 retweeted
love the “capability gaslighting” chat in this talk. frontier models would convince dawgs they’re an expert at some but fail the same task the next day.
What can us, engineers else learn from designers and design engineers? Turns out there’s lots, based on talking with one of the best design engineers in the industry, @Mappletons. Timestamps: 00:00 Intro 03:24 From anthropology to tech 10:18 What does a designer do? 18:23 How Maggie works 24:55 The case for planning with physical tools 31:53 Why Maggie is learning woodworking 33:13 Design engineers and engineering constraints 38:49 How Maggie uses Figma 40:30 Design at GitHub Next 45:12 How has AI changed design 50:37 When models design and why humans are still needed 53:30 UX and UI 58:29 Capability gaslighting 1:00:33 One Developer, Two Dozen Agents, Zero Alignment 1:07:21 Craft and AI tells 1:14:17 Visual gardens, home-cooked software, and barefoot developers 1:21:02 Advice for engineers and lessons from anthropology 1:25:34 Book recommendation Brought to you by: • @turbopuffer – a vector and full-text search engine built on object storage. It’s fast, cheap, and extremely scalable. turbopuffer.com/pragmatic • @AntithesisHQ – verify your system’s correctness without human review or traditional integration tests – and avoid bugs or outages. antithesis.com/pragmatic • @EntireHQ – every agent prompt, tool call, stored in your repo, and mirrored. Set up with one click: entire.io/pragmatic Hope you enjoy this unusually visual episode - thanks again, Maggie!
1
2
4
175
Replying to @Helius
@Helius na una go later kill me one day too :| makachi
1
2
3
203
anger means i still care silence means i am done!🙂‍↕️
19
Vinyl Davyl 🪽 retweeted
Replying to @Helius
@Helius na una go later kill me one day too :| makachi
1
2
3
203
Vinyl Davyl 🪽 retweeted
until you ask how much complexity they added to save them. performance is a numbers game. tops
100 TB of RAM, saved by shrinking a consistent hash ring. The last 90,000 hashes per server were buying 0.7% load balance improvement. Math said stop. We stopped. blog.cloudflare.com/saving-1…
2
3
196
Vinyl Davyl 🪽 retweeted
i think we’re going to look back at “stream the LLM tokens into a chat bubble” as the first primitive of AI interfaces real agent interfaces need to stream much more than text, if the agents aren’t streaming then i don’t know what that is
1
2
4
145
Vinyl Davyl 🪽 retweeted
inspired by @ephraimduncan would be having a guest book of signatures on my wall. live the test of time, why not
🚧 new overdue minimal corner of the web launching sometime this week
2
10
455